Demand for low-power data processing hardware continues to rise inexorably. Existing programmable and "general purpose" solutions (eg. SIMD, GPGPUs) are insufficient, as evidenced by the order-of-magnitude improvements and industry adoption of application and domain-specific accelerators in important areas like machine learning, computer vision and big data. The stark tradeoffs between efficiency and generality at these two extremes poses a difficult question: how could domain-specific hardware efficiency be achieved without domain-specific hardware solutions? In this work, we rely on the insight that "acceleratable" algorithms have broad common properties: high computational intensity with long phases, simple control patterns and dependences, and simple streaming memory access and reuse patterns. We define a general architecture (a hardware-software interface) which can more efficiently expresses program with these properties called stream-dataflow. The dataflow component of this architecture enables high concurrency, and the stream component enables communication and coordination at very-low power and area overhead. This paper explores the hardware and software implications, describes its detailed microarchitecture, and evaluates an implementation. Compared to a state-of-the-art domain specific accelerator (DianNao), and fixed-function accelerators for MachSuite, Softbrain can match their performance with only 2x power overhead on average.

General purpose processors (GPPs), from small inorder designs to many-issue out-of-order, incur large power overheads which must be addressed for future technology generations. Major sources of overhead include structures which dynamically extract the data-dependence graph or maintain precise state. Considering irregular workloads, current specialization approaches either heavily curtail performance, or provide simply too little benefit. Interestingly, well known explicit-dataflow architectures eliminate these overheads by directly executing the data-dependence graph and eschewing instruction-precise recoverability. However, even after decades of research, dataflow architectures have yet to come into prominence as a solution. We attribute this to a lack of effective control speculation and the latency overhead of explicit communication, which is crippling for certain codes.This paper makes the observation that if both out-of-order and explicit-dataflow were available in one processor, many types of GPP cores can benefit from dynamically switching during certain phases of an application's lifetime. Analysis reveals that an ideal explicit-dataflow engine could be profitable for more than half of instructions, providing significant performance and energy improvements. The challenge is to achieve these benefits without introducing excess hardware complexity. To this end, we propose the Specialization Engine for Explicit-Dataflow (SEED). Integrated with an inorder core, we see 1.67× performance and 1.65× energy benefits, with an Out-Of-Order (OOO) dual-issue core we see 1.33× and 1.70×, and with a quad-issue OOO, 1.14× and 1.54×.